1. Field of the Invention
The invention relates generally to a method and apparatus for purifying target biological substance(s), such as selected proteins, antibodies, antigens, clotting factors, glycoproteins, and hormones, from source liquids containing contaminants that have molecular weights or other physical or chemical properties that differ from those of the target substance, wherein the purification is effected by sequential chromatographic and diafiltration separation steps in a cross-flow filtration system.
2. Description of the Related Art
Various methods of purification have been employed for the separation of substances from liquid samples. Precipitation, centrifugation, filtration, chromatography and evaporation have all been employed with varying success with respect to yield, time consumption, purity and cost.
In the area of biological purification, centrifugation, chromatography and filtration have been especially useful for obtaining highly valuables substances from liquid samples with yields ranging from 10 to 90 percent and purity as high as 95 percent.
In current applications of centrifugation, chromatography and filtration, it is generally understood that yield and purity are in an inverse relationship and that yields are significantly lower for each subsequent purification step. It is also well understood that these methods of centrifugation, chromatography and filtration are expensive, relatively slow, and employ equipment that is very difficult to clean prior to its reuse.
A particular problem in this respect is the cleaning of fixed bed chromatography columns, in which irregular flow channels tend to be formed through the chromatography resin. These irregular flow channels present a particular problem in the purification of biological substances, since a failure to completely clean the column can result in the contamination of subsequent batches.
An example is the purification of plasma proteins on ion exchange and affinity chromatography columns. If a batch of plasma tested to be free of virus is later learned to be contaminated with virus, it is nearly impossible to calculate the certainty of removal of the virus from the column. In addition, because biological liquids readily support the growth of bacteria, simple bacterial contamination and growth of organisms in chromatographic columns is by no means infrequent. Bacterial organisms and the endotoxins produced by the bacteria have contaminated countless batches of pharmaceutical products resulting in significant financial losses as well as adverse reactions in the recipients of the final product.
The frequently observed "rat tunnels" which present so many problems for validation of the cleaning process also negate a significant portion of the capacity and resolution capability of chromatography columns.
Another problem of fixed bed chromatography columns is compression of the resin, particularly in the case of softer gels such as agarose (e.g., Sepharose.RTM. gel, commercially available from Pharmacia). The joint problems of tunneling and compression significantly raise the cost of chromatography by necessitating large amounts of excess binding capacity. Another problem caused by compression and tunneling is loss of purity. High purity requires uniform elution of the target substance. Tunneling and compression prevent uniform distribution of the elution liquid, resulting in imprecise separation of the target substance from contaminants which have similar elution profiles to the target product as well as to randomly eluted contaminants entrapped in the compressed media.
In the case of monoclonal antibody purification, it is a common practice to pack a column with a ten-fold excess binding capacity. In a well-distributed system it would be possible to bind the entire target product with only a three-fold excess capacity, thereby reducing the cost of the chromatography media three-fold.
One common approach to decreasing tunneling and compression is to lower the operating pressure of the column by reducing the flow rate. Although the practice of reducing the flow rate decreases the compression of the resin, it significantly increases the processing time and in many cases adversely effects the resolution and the yield of the process.
Tangential flow filtration utilizes membranes of various pore sizes for separating substances in liquids by pumping the liquid parallel to the membrane surface. Although this process has proven effective in the concentration of substances suspended in water and/or buffers, it has not proven widely useful in the purification of compounds in solution. The first problem of this method is that the pore size is not sufficiently uniform to allow for the separation of two closely sized particles. In addition, substances in the liquid mixture, especially proteins and lipids, bind to the surface of the membrane, a phenomenon referred to as "gel layer polarization," changing the effective pore size as well as the surface chemistry of the membrane.
Fluidized bed chromatography is another means of separating substances from liquid mixtures. Fluidized bed chromatography is more commonly utilized in the chemical and petroleum industries. Fluidized bed columns are frequently 10 feet high or higher and 9 to 12 inches in diameter. Pharmaceutical and bioprocess columns are usually less than 3 feet high and have a wide variety of diameters in the general range of from 1 to 24 inches, depending on the compression characteristics of the resin. The advantages of a fluidized bed are higher flow rates at lower pressures as compared to fixed bed chromatography. Although the higher flow rates offer certain advantages to the chromatographic separation, the method has several shortcomings. The method requires larger diameter resins that are neutral to gravity or buoyant. These larger, 100 to 300 micron mean diameter resins have less surface area per unit volume than smaller, 1 to 100 micron resins used in fixed bed columns, and correspondingly have less surface binding capacity.
To minimize the loss of surface area and decrease density, the fluidized bed resins are highly porous structures. These resin particles, however, as a result of their porous character, are highly susceptible to cracking, thereby generating small particulates that block the inlet and outlet ports of the column.
The most significant problem of the fluidized bed is mixing. Since the column does not contain any static mixing means, the bed is conventionally mixed by means of air jets or by recycling the liquid to be separated through the column at a high flow rate. The high flow rate and limited mixing inhibit the uniform phase change required during elution of the product from the resin.
As a result of the above-described deficiencies in the art, there is a compelling need for a rapid, uniform, time- and cost-efficient system for purifying biological target substances from complex liquid sources. Such a system would desirably overcome the problems inherent in the various prior art separation technologies described above. Such a system also would desirably be readily scalable, being adaptable to process volumes of source material ranging from milliliters in the research laboratory to the thousands of liters commonly encountered in biopharmaceutical production. Finally, such a system would desirably be capable of use with source liquids of widely varying properties, including viscous complex solutions.